Understanding how neurons code and process information continues to be a major goal of neuroscientists. Better understanding of neural coding would have wide clinical benefits, for example, in the design of neural implants. Many researchers are recording the electrical spike trains of single neurons and ensembles of neurons to learn about neural coding. Currently available commercial software for analysis of single- unit and multi-unit data is not adequate, especially for the increased demands of analyzing neural ensemble recordings. It is clear that the technology to obtain multi-unit data has far outstripped the currently available methods to analyze such data. The long-term objective of this research program is to develop a MATLAB toolbox for the analysis of single and ensemble neural spike train data, based on advanced signal processing and information theoretic algorithms. The proposed project will produce the first commercially available software to consolidate standard and novel neural analysis techniques for both single-unit and multi-unit neural data. The objective of the Phase I project is to develop a prototype software environment in MATLAB and to research and develop general purpose algorithms for the analysis of single-unit and multi-unit data. PROPOSED COMMERCIAL APPLICATIONS: There are many potential users of spike analysis software, including those researchers doing single-unit and multi-unit recordings. The number of researchers using multi-channel recordings has increased and continues to increase as computers become faster and better multi- electrodes are produced. Spike analysis software that includes multi- unit techniques is currently in demand from this community and will continue to be in demand.